The broad goals of the work are to significantly improve real-time fluoroscopic images while reducing the patient's radiation dose. The research may show that, by redistributing the x-ray photon fluence so that more photons are sent to a region of interest (ROI) and fewer to a region peripheral to the ROI, an improved image may be obtained in the region of clinical activity at a lower integral-dose. The specific clinical applications for the proposed research are interventional radiography and pediatric fluoroscopy. For the first, the aim is to evaluate how much better, and with what integral dose reduction a given interventional task can be accomplished if the real-time image in an ROI where the interventional action is occurring (i.e. near the catheter tip) is improved with the sacrifice of a degraded or noisier image in the periphery. Additionally, with a small ROI, small rotations of the x-ray beam about the treatment volume can spread the radiation over the entrance surface so that the common radiation induced clinical problems of skin reactions and hair loss may be eliminated without impacting the interventional procedure. For the second application, the aim is to evaluate the pediatric dose savings achievable using ROI fluoroscopy with no significant change in the diagnostic utility of the procedure. Specific goals include 1) construction of an ROI filter mechanism and 1000-line real-time digital subtraction system; 2) construction of vascular phantoms allowing evaluation of simulated interventional tasks for task accomplishment, duration of procedure, and integral and entrance dose saving with and without dose rate increase in the ROI, as a function of ROI filter parameters; 3) evaluation of ROI fluoroscopy for interventional tasks in a canine aneurysm model; and finally 4) a multi- site clinical study of ROI fluoroscopy. In order to determine the efficacy of ROI fluoroscopy, of necessity, the image quality requirements for clinical fluoroscopy and, in particular, for conducting interventional radiologic tasks will be analyzed. Ultimately, it is hoped that the research will lead to better utilization of ionizing radiation in medical diagnosis and intervention.